Assessing concordance between patient recall and health record documentation of medication discussions in Veterans Affairs primary care: A brief report

Abstract: Objective: High quality patient-clinician communication is critical for patient-centered care, particularly for medication management and discussions. This study assessed concordance between patient report and electronic health record (EHR) documentation of medication discussions following a patient-engagement deprescribing intervention. Methods: This secondary analysis used data from a pragmatic clinical trial in which 3,206 Veterans Affairs patients received a mailed educational intervention about deprescribing proton pump inhibitors, high-dose gabapentin, or insulin/sulfonylureas before scheduled primary care visits. We assessed patient-reported discussions using post-appointment surveys. We reviewed 72 EHR notes for evidence of medication discussions. Chi-square tests analyzed concordance between patient report and EHR documentation. Results: Among 72 sampled participants, 43% reported a medication discussion with their primary care practitioner, while 44% of charts contained documentation of discussions. Most patients (81%) who reported a discussion indicated they initiated it. There were 45 concordant cases between patient-report and chart documentation (62.5%); 27 where neither reported a discussion and 18 where both indicated a discussion. There were 27 discordant cases (37.5%); in 13 only the patient reported a discussion and in 14 only the EHR contained documentation. Clinical records never explicitly indicated the initiator. The observed differences between patient self-report and EHR documentation were statistically significant (p = 0.043). Conclusion: Medication discussions occurred in nearly half of cases, supporting the effectiveness of educational brochures in fostering patient-clinician discussions. However, the presence of 37.5% discordance between patient recall and EHR documentation underscores the need for communication strategies that ensure complete, understandable transmission of information. Most (81%) patients who reported a discussion reported initiating it, suggesting that the educational intervention engaged patients in their care. Practice Implications Providing patients with educational materials can promote agency and involvement of patients in their own care. Ensuring that discussions are high-quality and patient-centered may increase concordance between patient recall and EHR documentation.